Waveguide twists are used to rotate the field orientation for matching two waveguides exhibiting an angular offset. In solutions known in the art the vector of the electric field is rotated in intermediate waveguide sections with appropriate angular steps from the input to the output waveguide. Each angular step gives rise to a partial reflection of the wave depending on the angular increment. In a proper design, these partial reflections should cancel at the center frequency; therefore the length of each section is favorably in the order of a quarter waveguide wavelength (or an odd multiple thereof). The overall bandwidth depends on the number of waveguide sections.
State-of-the-art waveguide twists are commonly based on step-twist sections, as is introduced, for example, in Wheeler, H. A., et al., “Step-twist waveguide components”, IRE Trans. Microwave Theory Tech., vol. MTT-3, pp. 44-52, October 1955. To adapt the interconnection of two interface waveguides with a T-shape alignment (i.e. 90-degree angular offset), this solution can be modified considering, in addition to the angular offset between the intermediate steps, an offset along the cross section axis of one of the interfacing waveguides. A suitable realization of this design in one piece is possible by machining the structure from the flange faces with state-of-the-art CNC milling techniques. However, such a design is only possible for no more than two transformer steps, which yields substantial limitations for the achievable performance (i.e., Voltage Standing Wave Ratio, VSWR, and bandwidth). The length of the component is determined by the frequency band, i.e. the length of each transformer step is at a quarter waveguide wavelength of the center frequency of the operating band. Another drawback of the prior art solutions results from the fact that this solution would commonly exhibit an angular offset at the flange interconnections (interfaces). In consequence a specific (i.e. non-standard) flange sealing is necessary when using this component in sealed (pressurized) waveguide systems.
Alternative solutions known in the art are those consisting of two parts that have to be connected to form a fully functional junction. The two part format of these junctions allows for more complicated machining and, in consequence, achieves improved performance. However, the manufacturing of such junctions is complicated, expensive and time consuming. If two (or more) parts are used, they need to be combined in an appropriate way, which increases the manufacturing effort and expense. They could be assembled by screws—but such a solution needs additional sealing means in the parting plane if the component is used in a pressurized waveguide system. Another approach could entail assembling the parts by soldering or brazing—however, such solutions need the careful choice of the basic (and surface) material and the overall construction to accommodate the requirements of the additional process. Moreover the realization of the component from two (or more) parts yields additional tolerances (e.g., fitting of the parts) that may impair the optimal performance.
Another solution known in the art is the one defined in U.S. Pat. No. 6,756,861. Such a solution would allow the interfacing of orthogonally aligned waveguides with arbitrary offsets. But for a T-shape structure an additional bend has to be integrated into the structure, which increases the size and the unit becomes bulky. It should also be noted, that such a solution in general requires that the twist consists of two parts.
Hence, an improved waveguide junction would be advantageous and in particular one that has good performance characteristics and is easy for manufacturing.